Despite proven pharmacological efficacy, naltrexone has had little use in treating opioid dependence because of poor patient compliance, the problem we address in this proposed project. In the first study of Behavioral Family Counseling (BFC) and naltrexone, we randomly assigned male opioid dependent patients who were living with a family member to equally intensive treatments: (a) BFC with a family member observing the patient take naltrexone daily plus individual-based treatment (IBT) or (b) IBT only. BFC patients, compared with their IBT counterparts, ingested more doses of naltrexone, attended more treatment sessions, stayed abstinent longer, and had significantly more days abstinent from opioids and other illicit drugs during treatment and in the year after treatment. BFC patients also had significantly fewer drug related legal and family problems at 1-year follow-up. The proposed project will conduct a more comprehensive randomized clinical trial comparing 3 interventions each of which includes a prescription for naltrexone (50mg/day) with a standardized compliance enhancement method plus individual drug counseling (IDC). Opioid dependent patients (N=225) seeking outpatient drug free treatment (generally immediately post-detoxification) will be randomly assigned to 24 treatment sessions over a 16-week period consisting of either: (a) behavioral family counseling in which a family member observes the patient's daily naltrexone ingestion (BFC-NAL) plus IDC;(b) behavioral family counseling (BFC) without family observation of naltrexone ingestion plus IDC;or (c) IDC for the patient alone. We will test the prediction that both family treatment conditions will do better than IDC alone and that family counseling with daily observed naltrexone will do better than family counseling without such observation (i.e. BFC-NAL >BFC >IDC) on (a) primary outcomes of greater abstinence from opioids;(b) intermediate outcomes of greater naltrexone compliance and longer retention in treatment;and (c) secondary outcomes of more abstinence from drugs other than opioids, more positive family and psychosocial functioning, and fewer HIV risk behaviors. Outcome data will be collected from patients and family member collaterals during treatment, immediately posttreatment, and at quarterly follow-ups for 12 months after treatment.